基于模糊关联规则挖掘的网络入侵检测算法

发布时间：2018-02-08 13:30

本文关键词： 网络安全入侵检测关联规则数据挖掘　出处：《现代电子技术》2017年09期 　论文类型：期刊论文

【摘要】：为了有效解决当前网络入侵检测算法存在的缺陷,提高网络的安全性,提出基于模糊关联规则挖掘的网络入侵检测算法。首先收集网络数据,提取网络入侵行为的特征;然后采用模糊关联规则算法对入侵行为特征进行挖掘,选择入侵行为最有效的特征,减少特征之间的关联度;最后支持向量机根据"一对多"的思想建立网络入侵检测的分类器,以KDD CUP数据为例对网络入侵检测性能进行分析。结果表明,该算法的网络入侵检测正确率超过了95%,检测结果要明显好于其他检测算法,易实现,可以用于大规模网络的在线入侵检测分析。
[Abstract]:In order to effectively solve the shortcomings of current network intrusion detection algorithms and improve the network security, a network intrusion detection algorithm based on fuzzy association rule mining is proposed. Firstly, the network data is collected to extract the characteristics of network intrusion behavior. Then the fuzzy association rule algorithm is used to mine the intrusion behavior feature, and the most effective feature is selected to reduce the correlation degree between the features. Finally, support vector machine (SVM) constructs a classifier for network intrusion detection based on the idea of "one-to-many". Taking KDD CUP data as an example, the performance of network intrusion detection is analyzed. The results show that, The network intrusion detection accuracy of this algorithm is over 95, and the detection result is obviously better than other detection algorithms. It is easy to implement and can be used for on-line intrusion detection and analysis of large-scale networks.
【作者单位】：百色学院信息工程学院;
【分类号】：TP393.08;TP311.13

【相似文献】